Online Learning of Approximate Dependency Parsing Algorithms

By Ryan T. McDonald, Fernando C Pereira. Conference of the European Chapter of the Association for Computational Linguistics, 2006

562 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/8d1d98807843fad7de1734629edb0c873015c14a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Dependency-based Convolutional Neural Networks for Sentence Embedding (2015, methodology): “Inspired by higher-order dependency parsing (McDonald and Pereira, 2006; Koo and Collins, 2010), we also incorporate siblings for a given word in various ways.”
Structured Prediction Models via the Matrix-Tree Theorem (2007, background): “For higher-order non-projective parsing, however, computational complexity results (McDonald and Pereira, 2006; McDonald and Satta, 2007) indicate that exact solutions to the three inference problems of Section 2.2 will be intractable.”
Multilingual Dependency Learning: Exploiting Rich Features for Tagging Syntactic and Semantic Dependencies (2009, methodology): “Basically, we build our syntactic dependency parsers based on the MSTParser, a freely available implementation 2 , whose details are presented in the paper of McDonald and Pereira (2006).”
Extended Constituent-to-Dependency Conversion for English (2007, background): “A further step could be to extend the dependency structures to allow multiple-headed graphs, for which a practical parsing algorithm was recently proposed (McDonald and Pereira, 2006).”
Syntactic Processing Using the Generalized Perceptron and Beam Search (2011, methodology): “Templates 1–6 are taken from MSTParser (McDonald and Pereira 2006), a second-order graph-based parser.”
A Discriminative Graph-Based Parser for the Abstract Meaning Representation (2014, background): “Mcdonald and Pereira (2006) incorporated second-order features, but resorted to an approximate algorithm.”
Transition-based Dependency Parsing with Rich Non-local Features (2011, methodology): “Compared to graph-based dependency parsing, it typically offers linear time complexity and the comparative freedom to define non-local features, as exemplified by the comparison between MaltParser and MSTParser (Nivre et al., 2006b; McDonald et al., 2005; McDonald and Nivre, 2007).”
Distributed Training Strategies for the Structured Perceptron (2010, methodology): “We used the CoNLL-X training (72703 sentences) and testing splits (365 sentences) of this data (Buchholz and Marsi, 2006) and dependency parsing models based on McDonald and Pereira (2006) which factors features over pairs of dependency arcs in a tree.”
Top Accuracy and Fast Dependency Parsing is not a Contradiction (2010, methodology): “It consists of the second order parsing algorithm of Carreras (2007), the non-projective approximation algo-rithm (McDonald and Pereira, 2006), the passive-aggressive support vector machine, and a feature extraction component.”
Dependency Parsing by Belief Propagation (2008, methodology): “Finally, Table 2 compares loopy BP to a previously proposed “hill-climbing” method for approximate inference in non-projective parsing McDonald and Pereira (2006). Hill-climbing decodes our richest non-projective model by finding the best projective parse under that model—using slow, higherorder DP—”
Vine Pruning for Efficient Multi-Pass Dependency Parsing (2012, unknown): “For second-order models with index set I1,0, parsing can be done in O(n3) time (McDonald and Pereira, 2006) and for third-order models in O(n4) time (Koo and Collins, 2010).”
An Efficient Algorithm for Easy-First Non-Directional Dependency Parsing (2010, background): “(McDonald and Pereira, 2006; Carreras, 2007)).”
Stack-Pointer Networks for Dependency Parsing (2018, methodology): “In this paper, we adopt the inside-out order3 since it enables us to utilize second-order sibling information, which has been proven beneficial for parsing performance (McDonald and Pereira, 2006; Koo and Collins, 2010) (see § 3.”
Dependency-based Semantic Role Labeling of PropBank (2008, methodology): “…of the complete parse tree by summing the scores of its parts, referred to as factors: Ψ(x, y) ·w = ∑ f∈y ψ(x, f) ·w We used a second-order factorization (McDonald and Pereira, 2006; Carreras, 2007), meaning that the factors are subtrees consisting of four links: the governor–dependent link, its…”